Recombinant DNA Advisory Committee - 12/3-4/92 
denuded of its epithelial cells. The human-seeded rat trachea is then grown in the nude 
mouse. The CF mouse model includes tubing that has been tied proximally and distally 
so that it imitates an airway, including mucus production. This airway must be irrigated 
several times a day in order to prevent occlusion. The advantage of this model is that it 
provides a setting for the introduction of genes into human airway epithelial cells. 
He described in vivo experiments in which lac/Z and CFTR were injected into the lumen 
of the CF mouse. The human xenografts were subsequently removed and evaluated for 
gene expression. The epithelium generated on the surface of the xenograft is identical to 
that of the proximal airway, i.e., ciliated cells, goblet cells, and basal cells. Data 
demonstrates that 10 12 PFU injected into the lumen results in a graft transduction 
efficiency between 15 and 40%. Injection of 10 9 PFU resulted in transduction of 1 to 4% 
of the graft. When the human xenografts were left in the animals for 3 weeks following 
transduction, no diminution of gene expression was observed during this period of time. 
Dr. Wilson presented data regarding the types of cells that are infected by the vector. 
Using immunofluorescent staining, he demonstrated that there is co-localization of the 
recombinant gene protein only in differentiated cells, not progenitor cells. Consequently, 
it may be necessary to administer subsequent vector challenges. 
The human xenografts have been evaluated for adenovirus protein expression. Double 
immunofluorescence was used to determine which cells contain the recombinant 
adenovirus and if these cells express the adenovirus protein. Data suggests that there is 
immune localization to a 72 kd early protein, E2a. Between 3 and 5% of the human 
cells express E2a. The same experiment was performed to determine expression of a 
late gene product using exon and fiber protein antibody. Expression of these late gene 
products were essentially undetectable. This result is consistent with other data 
regarding El-deleted adenoviruses. Specifically, there is a block in the transition from 
early to late protein expression. 
Dr. Wilson presented virus distribution experiments that were performed in the baboon 
animal model. Animals received 20 ml of lac/Z virus into the posterior segment of the 
left upper lobe. The lung was removed on day 3, perfused, and washed with X-gal to 
evaluate lac/Z expression in situ. Results indicate that the majority of gene expression 
remains in the posterior segment. The reason for localizing the infusion of the virus is to 
minimize the amount of lung at risk in the event that adverse reactions occur. 
Dr. Wilson described an in vivo toxicity experiment in which a 33 kilogram baboon 
received the CFTR adenovirus into the posterior segment of the right upper lobe and 
lac/Z into the left upper lobe. Bronchial lavage was performed on both lobes on days 3, 
14, and 21, in addition to brushings and biopsies. Although the data analysis is still in 
progress on the samples obtained from this animal, preliminary data indicates a variation 
Recombinant DNA Research, Volume 16 
[567] 
